Spinal cord injury (SCI) disrupts connections between the brain and spinal cord, causing devastating loss of mobility and independence. Most injuries are incomplete (iSCI), leaving intact at least some neural pathways to motor neurons that control movement. Although spontaneous plasticity in these spared pathways underlies some functional recovery, the extent of spontaneous recovery after iSCI is slow, variable and frustratingly limited. There is a critical need for new therapies that induce further improvements in persons with chronic iSCI. We recently demonstrated that repetitive exposure to acute intermittent hypoxia (rAIH), alone or in combination with walking training, stimulates motor recovery in persons with chronic iSCI. As an early stage investigator, I now propose to test four hypotheses concerning mechanisms of rAIH-induced motor recovery using multiple experimental approaches, including muscle electromyography, measurements of walking dynamics and leg strength, genotyping and pharmacological intervention. In Aim 1, we will focus on the neuromechanical bases of improved walking after rAIH, including increased motor gain (muscle activity) and coordination, as well as, reduced step-by-step variability. In Aim 2, we will test whether the functional benefits of rAIH when combined with training (strength or walking) are task-specific. We hypothesize that the cellular mechanisms that underlie the benefits of rAIH (alone or combined with training) in respiratory and non- respiratory motor function in rodent models applies also to humans with chronic, iSCI. Since AIH induces spinal motor plasticity in rats by a mechanism that requires serotonin-dependent synthesis of brain derived neurotrophic factor (BDNF), in Aim 3, we will explore the BDNF-dependence of rAIH-induced motor recovery by assessing the impact of extent of functional recovery in individuals with bdnf polymorphisms known to undermine spinal BDNF function in humans. Finally, concurrent activation of competing cellular cascades during AIH (initiated by serotonin and adenosine, respectively) undermines respiratory motor plasticity in rats. By removing the adenosine-dependent mechanism, greater functional plasticity can be achieved. Thus, in Aim 4, we hypothesize that, by pharmaceutically removing the adenosine constraint on rAIH-induced spinal plasticity via caffeine administration, the therapeutic efficacy of rAIH will improve, increasing its potential asa viable treatment to improve motor function. Each aim is supported by substantial preliminary data, suggesting that the proposed experiments will advance our understanding of mechanisms giving rise to rAIH-induced motor recovery after iSCI. An important goal guiding our research is to identify ways to optimize rAIH-induced plasticity, thereby promoting meaningful functional recovery in persons with chronic iSCI.